PERF02-BP06 Use optimized hardware-based compute accelerators

Use hardware accelerators to perform certain functions more efficiently than CPU-based alternatives.

Common anti-patterns:

In your workload, you haven't benchmarked a general-purpose instance against a purpose-built instance that can deliver higher performance and lower cost.
You are using hardware-based compute accelerators for tasks that can be more efficient using CPU-based alternatives.
You are not monitoring GPU usage.

Benefits of establishing this best practice: By using hardware-based accelerators, such as graphics processing units (GPUs) and field programmable gate arrays (FPGAs), you can perform certain processing functions more efficiently.

Level of risk exposed if this best practice is not established: Medium

Implementation guidance

Accelerated computing instances provide access to hardware-based compute accelerators such as GPUs and FPGAs. These hardware accelerators perform certain functions like graphic processing or data pattern matching more efficiently than CPU-based alternatives. Many accelerated workloads, such as rendering, transcoding, and machine learning, are highly variable in terms of resource usage. Only run this hardware for the time needed, and decommission them with automation when not required to improve overall performance efficiency.

Implementation steps

Identify which accelerated computing instances can address your requirements.
For machine learning workloads, take advantage of purpose-built hardware that is specific to your workload, such as AWS Trainium, AWS Inferentia, and Amazon EC2 DL1. AWS Inferentia instances such as Inf2 instances offer up to 50% better performance/watt over comparable Amazon EC2 instances.
Collect usage metrics for your accelerated computing instances. For example, you can use CloudWatch agent to collect metrics such as utilization_gpu and utilization_memory for your GPUs as shown in Collect NVIDIA GPU metrics with Amazon CloudWatch.
Optimize the code, network operation, and settings of hardware accelerators to make sure that underlying hardware is fully utilized.
Use the latest high performant libraries and GPU drivers.
Use automation to release GPU instances when not in use.

Resources

Related documents:

Related videos:

Related examples:

Warning Javascript is disabled or is unavailable in your browser.

To use the Amazon Web Services Documentation, Javascript must be enabled. Please refer to your browser's Help pages for instructions.

Document Conventions

PERF02-BP05 Scale your compute resources dynamically

Data management